Improved energy production of wind turbine parks

Use Case Description

Goal of the use case

Producing green energy is an expensive process. In the eight offshore parks of the Belgian coast, there are about 400 wind turbines that supply energy to 2.2 million families. Keeping this infrastructure operational is not only extremely expensive, but also crucial for a stable power supply. The modern wind turbines are therefore equipped with about 800 sensors that transmit data every second. The data is interpreted in real-time, on the one hand to monitor the condition of each machine and to be able to plan the necessary maintenance work, and on the other hand also to improve energy production.

The research within the Flanders AI Research Program

Thanks to AI, wind turbines can inform the database and each other to adapt to real weather conditions. In this way, AI can ensure that the wind turbines are not overloaded and that production is more efficient.

Results-Artefacts

Demonstrator

Wind turbine demo

Journal/Conference publications

Perez Sanjines, F. R., Peeters, C., Verstraeten, T., Vervlimmeren, I., Nowe, A., & Helsen, J. (2023). Early Fault Detection Using Deep Learning on Compressed Cyclic Spectral Coherence Maps. In ASME Turbo Expo 2023: Turbomachinery Technical Conference and Exposition (Vol. 14). [v014t37a001] (Proceedings of the ASME Turbo Expo; Vol. 14). ASME. 

Chesterman, X., Verstraeten, T., Daems, P-J., Nowe, A., & Helsen, J. (2023). Pattern mining based data fusion for wind turbine condition monitoring. In Journal of Physics: Conference Series (Vol. 2507). IOP Publishing. https://doi.org/10.1088/1742-6596/2507/1/012001

Perez Sanjines, F. R., Peeters, C., Verstraeten, T., Antoni, J., Nowe, A., & Helsen, J. (2023). Fleet-based early fault detection of wind turbine gearboxes using physics-informed deep learning based on cyclic spectral coherence. Mechanical Systems and Signal Processing, 185, [109760]. https://doi.org/10.1016/j.ymssp.2022.109760

Chesterman, X., Verstraeten, T., Daems, P-J., Nowe, A., & Helsen, J. (2023). Overview of normal behavior modeling approaches for SCADA-based wind turbine condition monitoring demonstrated on data from operational wind farms. Wind Energy Science, 8(6), 893–924. https://doi.org/10.5194/wes-8-893-2023

Perez Sanjines, Fabian Ramiro; Peeters, Cédric; Verstraeten, Timothy; Antoni, Jérôme; Nowe, Ann; Helsen, Jan. Fleet-based early fault detection of wind turbine gearboxes using physics-informed deep learning based on cyclic spectral coherence. In: Mechanical Systems and Signal Processing, Vol. 185, 15.02.2023.

X. Chesterman, T. Verstraeten, PJ. Daems, A. Nowé and J.Helsen, "Condition monitoring of wind turbines and extraction of healthy training data using an ensemble of advanced statistical anomaly detection models", Proceedings of the Annual conference of the PHM Society, Vol. 13, No. 1.

Chesterman, Xavier; Verstraeten, Timothy; Daems, Pieter-Jan; Perez Sanjines, Fabian Ramiro; Nowe, Ann; Helsen, Jan. The detection of generator bearing failures on wind turbines using machine learning based anomaly detection. Journal of Physics Conference Series . Vol. 2265 3. ed. IOP Publishing, 2022. (Journal of Physics: Conference Series).

Perez Sanjines, Fabian Ramiro; Verstraeten, Timothy; Nowe, Ann; Helsen, Jan. Deep ensemble with Neural Networks to model power-curve uncertainty. Journal of Physics: Conference Series: 17th EERA DeepWind 2022 offshore wind R&I conference 19–21 January 2022, Trondheim, Norway. 2022.

Chesterman, X., Verstraeten, T., Daems, P-J., Nowe, A., & Helsen, J. (2021). Condition Monitoring of Wind Turbines and Extraction of Healthy Training Data Using An Ensemble of Advanced Statistical Anomaly Detection Models. In Proceedings of the Annual Conference of the PHM Society PHM Society. 

PhDs

Xavier Chesterman, Fabian Perez-Sanjines

Uptake

Related or Follow-up projects 

• The SuperSized 4.0 project aims to create an integrated methodology for wind turbines capable of assessing the risks for long-term underperformance, downtime due to failure and excessive lifetime consumption.

• The H2020 EU-funded PLATOON project provides new approaches and analytics tools for Energy Big Data, thus supporting the zero-carbon transition and developing new services in the energy domain. In an increasingly complex and heterogeneous environment, PLATOON enables the evolution from a classical centralized energy sector to a more distributed one, with intermittent renewable energy sources and new extended digital capabilities. While contributing to artificial intelligence, interoperability, data privacy and security, PLATOON adheres to the standards of the International Data Spaces Association (IDSA), aiming thus to realize the first IDS-compliant Data Marketplace for the Energy sector.

• The MaSiWEC project aims to create a synergetic framework for “Material Based Probability of Failure (MB-PoF)”. This framework is validated and strengthened by an on-shore turbine measurement campaign executed by NREL and a measurement campaign in a Flemish off-shore wind-mill park.

• The Rainbow project will provide the offshore wind energy industry with an improved understanding of the drivers for leading edge erosion (LEE) of the offshore wind turbine blades, resulting in new predictive maintenance capabilities, and optimized strategies for inspection, maintenance & repair.

• Position paper on state-of-the-art technologies for wind turbine drivetrains in collaboration with multiple universities of the European Academy for Wind Energy (EAWE).

• Article on Digitale Toekomst

Basic Use Case Info

Period

2019-2023 

Contact

UC Leads:  Prof. Jan Helsen (VUB AVRG), Prof. Ann Nowé (VUB AI Lab)

Research Groups

VUB Acoustics & Vibrations Research Group, VUB Artificial Intelligence Lab